© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 20274
    Prev
    Next
    INDEX
    Explanations

    recipe

    np_max-act · gemini-2.0-flash

    It activates on passages that present a cooking recipe — i.e., ingredient lists, instructions, and recipe-format content.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    掌控
    -0.07
     locals
    -0.07
    ɽ
    -0.07
     Đó
    -0.07
     Gy
    -0.06
    快餐
    -0.06
     Dillon
    -0.06
     öğ
    -0.06
     Rum
    -0.06
    ,total
    -0.06
    POSITIVE LOGITS
    按规定
    0.07
    �
    0.07
     hourly
    0.06
    jet
    0.06
    anus
    0.06
     Arrow
    0.06
    <p
    0.06
     scissors
    0.06
    shall
    0.06
     signing
    0.06
    Activations Density 0.049%

    No Known Activations